Reshape dataframe in R, different dates

Question

I have data that looks like this:

ID	Name	Role	Status	Date
1	John	GM	Current	12.04.2021
1	Ann	GM	Previous	10.07.2020
1	Mary	GM	Previous	24.01.2017
2	Ann	GM	Current	12.04.2021
2	Josef	GM	Previous	02.07.2015
3	Sophie	GM	Current	12.04.2021
4	Ben	GM	Current	12.04.2021
4	Lucas	GM	Previous	30.07.2018
4	Peter	GM	Previous	18.04.2017
4	Susan	GM	Previous	16.09.2015

The ID is unique for each "business". First of all, I want to have 1 row for each ID, and then each date needs to generate a new column. The first date in each ID has to be in "Date1", second "Date2", etc. It's worth mentioning that my dataset can take different numbers of rows for each ID.

I will use this analysis to look at changes in General Managers (GM) for each business, therefore only interested in ID and Date.

My final dataset will look like as the table below:

ID	Date1	Date2	Date3	Date4
1	12.04.2021	10.07.2020	24.01.2017	NA
2	12.04.2021	02.07.2015	NA	NA
3	12.04.2021	NA	NA	NA
4	12.04.2021	30.07.2018	18.04.2017	16.09.2015

I have tried to search for previous cases with reshape in R studio, but have not found similar to mine. Can someone help me? Many thanks in advance!

Answer 1

Here is a data.table approach

library(data.table)

DT <- fread("ID     Name    Role    Status  Date
1   John    GM  Current     12.04.2021
1   Ann     GM  Previous    10.07.2020
1   Mary    GM  Previous    24.01.2017
2   Ann     GM  Current     12.04.2021
2   Josef   GM  Previous    02.07.2015
3   Sophie  GM  Current     12.04.2021
4   Ben     GM  Current     12.04.2021
4   Lucas   GM  Previous    30.07.2018
4   Peter   GM  Previous    18.04.2017
4   Susan   GM  Previous    16.09.2015")

# summarise dates by id
ans <- DT[, .(dates = paste0(Date, collapse = "#")), by = ID]
# now split
ans[, paste0("Date", 1:length(tstrsplit(ans$dates, "#"))) := 
      tstrsplit( dates, "#")][, dates := NULL][]

#    ID      Date1      Date2      Date3      Date4
# 1:  1 12.04.2021 10.07.2020 24.01.2017       <NA>
# 2:  2 12.04.2021 02.07.2015       <NA>       <NA>
# 3:  3 12.04.2021       <NA>       <NA>       <NA>
# 4:  4 12.04.2021 30.07.2018 18.04.2017 16.09.2015

Answer 2

Here's a tidyverse solution:

library(tidyverse)

df <- data.frame(
  ID = c(1, 1, 1, 2, 2),
  Name = c("John", "Ann", "Mary", "Ann", "Joseph"),
  Role = rep("GM", 5),
  Status = c("Current", "Previous", "Previous", "Current", "Previous"),
  Date = c("12.04.2021", "10.07.2020", "24.01.2017", "12.04.2021", "02.07.2015")
)

df

  ID   Name Role   Status       Date
1  1   John   GM  Current 12.04.2021
2  1    Ann   GM Previous 10.07.2020
3  1   Mary   GM Previous 24.01.2017
4  2    Ann   GM  Current 12.04.2021
5  2 Joseph   GM Previous 02.07.2015


dfnew <- df %>% 
  dplyr::group_by(ID) %>% 
  dplyr::mutate(rownum = row_number()) %>% 
  dplyr::select(ID, rownum, Date) %>% 
  tidyr::pivot_wider(names_from = rownum, values_from = Date, names_glue = "Date{rownum}")

dfnew

# A tibble: 2 x 4
# Groups:   ID [2]
     ID Date1      Date2      Date3     
  <dbl> <chr>      <chr>      <chr>     
1     1 12.04.2021 10.07.2020 24.01.2017
2     2 12.04.2021 02.07.2015 NA